Recent findings suggest that a significant proportion of the genetic variance in complex psychiatric disorders can be explained by a collection of individually rare, highly penetrant genetic variants. Each of these copy number variants (CNVs) may be a risk factor for disease, and many are non-recurrent and sporadic. Optimally, the identification of specific mutations would result in personalized treatment interventions tailored to the underlying biology of the mutation. We have identified a complex structural rearrangement at 9p24.1 that segregates with psychosis in one family. One gene in the rearranged region, glycine decarboxylase (GLDC), is involved in the degradation of glycine in glia cells and is triplicated in mutation carriers. Glycine is a co-agonist for the N-methyl-D-aspartate receptor (NMDAR). Carriers of the GLDC triplication would be expected to have low levels of brain glycine, resulting in NMDAR-mediated hypofunction, which has been strongly implicated in the pathophysiology of schizophrenia. The carriers of this mutation are strong candidates to benefit from glycine augmentation of their psychotropic drug regimens. One aim of this R21 application is to carry out a proof-of-principle double-blind placebo-controlled glycin augmentation trial in carriers of this mutation and to assess changes in clinical symptoms and neurocognitive function. We also propose to carry out targeted neurobiological follow-up of mutation carriers and non-carriers in the same family in order to characterize the brain structural, functional and neurochemical properties of this mutation. These studies will probe glycine homeostasis using proton magnetic resonance spectroscopy, assess the relationship between brain and plasma glycine levels following an acute oral dose of glycine, and examine brain pathways (magnocellular) implicated in dysregulation of NMDA-mediated neurotransmission. The same procedures (except for the glycine loading scans) will be re-administered to carriers after six weeks of open label glycine augmentation. The results will significantly enhance our understanding of the neurobiology of rare CNVs associated with psychosis and their relevance to disease pathophysiology. More importantly, the results will, for the first time, link pathophysiology and a medically actionable treatment intervention to underlying genetics, with potential benefit to other patients with neuropsychiatric disease who have mutations in either the same gene or in other genes/pathways that are impacted by the same or related aberrant biological processes.